A LVQ-based neural network anti-spam approach

Save this PDF as:
 WORD  PNG  TXT  JPG

Size: px
Start display at page:

Download "A LVQ-based neural network anti-spam email approach"

Transcription

1 A LVQ-based neural network anti-spam approach Zhan Chuan Lu Xianliang Hou Mengshu Zhou Xu College of Computer Science and Engineering of UEST of China, Chengdu, China Abstract: Along with wide application of nowadays, many spare s flood into people's inboxes and bring catastrophe to their study and work. This paper presents a novel anti-spare e- mail filter based-lvq network in terms of spare s which are mainly made up of several kinds commercial or political spare s at present. Our experiment has proved that the filter based on LVQ is superior to Bayes-based and BP-based approaches in total performances apparently. Keyword: LVQ, anti-spare filtering, mutual information, vector space model 1. Introduction Along with wide application of the Internet, has been used widely with its characteristics of high-speed, convenient, low cost and become an efficient and popular communication medium nowadays. However, a large number of spare s flood into people's mailboxes and bring catastrophes into their study and life. Sparn is annoying to most users, as they waste users time, money, network bandwidth as well as, meanwhile, clutter users' mailboxes, even be harmful, e.g. pornographic content. It was reported an American received 2200 pieces spare on average in Increasing by 2% per month, it will reach 3600 pieces spam s in A survey tl] by CNNIC found that every user in china received 13.7 piece s per week in 2004, including 7.9 piece spam s. In America, spare s make enterprises to be loss up to 9 billions per year [21. A study was reported that spam messages constituted approximately 60% of the incoming messages to a corporate network. Without appropriate counter-measures, the situation will continue worsening and spare will eventually undermine the usability of . In terms of content-based anti-spam filtering, s are usually regarded as particular texts. Cohen [31 used RIPPER algorithm to classify s. Sahami et al. [4] used Bayes theorem to filter spare s. It was proved that Bayes-based filtering approach outperforms keywords-based approach in performance. Xavier Carreras et al. t~j used Boosting algorithm to filter spare s. After testing public sample PU1 corpus, they found that their approach outperforms Bayes-based and decision tree-based methods. Duhong Chen et al. [6l compared four algorithms, Bayes, decision tree, neural networks, Boosting, and drew a conclusion that neural network algorithm has higher performance. James clark et al. [7] designed a 3 layers BP neural networks. It was shown that a BP network with IG has rather good effect of identifying spare in their experiment. This paper uses a LVQ network, which combines subclasses into a single class and forms complex class boundaries, to design an anti-spare neural network model and identify spare s which are mainly composed of commercial and political s. Experiments have proved that the LVQ-based anti-spare filter has better performance than Bayesbased and BP neural network.-based approaches. The remaining of this paper is organized as follows: section 2 introduces vector weight and feature extraction based on MI, section 3 introduces spare classification and describes anti-spare LVQ algorithm and parameters setting, Section 4 shows our experiments and results, finally, section 5 is some conclusions. 34

2 2. sample and data preprocessing 2.1 representation Vector space model is] is a text representing approach which is widely used and has good performance in TC. is regarded as a vector space which is composed of a group of orthogonal key words. Let the dimension of vector space be n, d represents by V (d) =(xt, x2, "", Xn ), with the value of elements of vector being weight of each feature key word in d.. We use TFIDF [9] approach to calculate feature weight. In TFIDF approach, the frequency of a key word in document is directly proportional to the frequency which the word appears in the document and is inversely proportional to the number of documents which contain the word. Therefore, TFIDF of word t i in document is (1): TFIDFi=TFi log (N/DFi) (1) Where, TFi is the frequency that word ti appears in document d, N are the total numbers of training documents, DFi represents the numbers of documents which contain word t i. 2.2 Feature extraction Feature of may be based on word (e.g. price, adult and shop) or phrase (e.g. on sale, be over 21) as well as non-textual properties (e.g. whether or not a mail contains attachments or html tag). We adopt feature based on word in our experiments in order to focus on our algorithm performance and simplify our test. A mail contains many different words. A large part of words contribute little to the classification, what's more, some words may play a negative role in classifying process. Hence, it is necessary to select some important words as features. We select features by MI tl01 (Mutual Information) method to compress features and reduce dimensionality. MI is widely used in text categorization. MI of word t corresponding to class is calculated (2): AxN Ml(t,s) ~ log (A + B)x (A + C) (2) where A are the numbers of s which contain word t and belong to class s, B are that of s which contain word but not belong to class s. C are that of s which belong to class s but not contain word t. N is the total number in training corpus. For multi-class, we compute MI of the word corresponding to every class respectively and then select the maximum according to formula (3), MZma x (t) = max,~ I MI(t, s,) (3) where m denote the numbers of classes, we will choose feature words whose MI are bigger than a threshold so as to decrease the dimensionality of vector space. 3. Anti-spam LVQ model 3.1 Spam category. Spare s, which mainly are commercial and political s at present, vary significantly in content. Related statistic [111 shows that spare s are mainly composed of shopping online, promoting IT products, get-rich, adult products, vacation, political 35

3 information, business information, pornography/violence as well as other in content, shown in figure others shopping 13% online 1, therefore, category of spare s is rather wide pornography/ 1 Clustering centers of different subclasses of spam s are business different, while feature words of infor: tion W 1 the whole spare s are sparse, so that it is difficult to political~ infor:~tion ~ distinguish spam s from legitimate s. If we divide avaction ~ 12% spam s into several 9% adult subclasses according to content, products 10% feature words of every subclass is closer and more related so as Fig.1 category of spam s to identify spare s easily. Hence, in our approach, if a mail belongs to one of subclasses of spare s, due to legitimate s being usually rather different from above subclasses of spam in content, the is regard as a spare Learning vector quantization neural network model LVQ network D21 is a hybrid network, which form classification through supervise and unsupervised learning. The model is divided into two layers. The first layer is competitive layer, in which each neuron represents a subclass, and the second is output layer, in which each neuron represents a class. A class may be composed of several subclasses. The second layer combines several subclasses into a class through W2 matrix. So LVQ network may create complex boundaries through combining several subclasses into a class. Therefore, LVQ is suited to classify spare s which have several subclasses. 3.3 Anti-spam LVQ algorithm initialize weight vectors W= { wl, w2, "", W n }, and learning rate ct ~ [0, 1] select an example from training corpus, and calculate distance between weight vectors and it respectively. We take the place of Euclidean distance in formula with Cosine distance, which represents similarity of two texts, cosine distance is defined as follows: Wu k *Wvk Sim (U,V) = k=j n 2 n 2 (4) then compare similarities between the example and each weight vector, in the result, the neuron with maximum similarity, wins and outputs 1, other neuron of hidden layer output 0 a' -- max(sire(x, x')) (5) 36

4 adjust weight, if a input example belongs to class r, the neuron c which wins in competitive learning belongs to class s, we will adjust weight in accordance with formula (6) f wc(t+l) = wc(t)+u(t)[x(t)-wc(t)]; r = s wo(t +1)= wc(t)-u(t)[x(t)-w~(t)]; r : : s (6) w i (t + 1) = wj (t); i ~ c modify learning rate u(t), decrease u(t) when iteration increasing check stopping condition, whether is iterative times enough. 3.4 Parameter setting Figure 2 shows our anti-spam LVQ network model, in input layer, we choose 100 feature words as input nodes. According to previous experiments, when the number of feature is set 100, the filter has a better performance, if the number continues to increase, the performance improves slightly, but calculation increases sharply. We choose 10 neurons in hidden layer with competitive function. On output layer, two neurons (legitimate and spam) and linear function are set. Input Competitive Layer Linear Layer t ntl-~'tm(x, x s) ~=compet (n 1) Fig2. Anti-spam s LVQ model a~=fa J J In order to accelerate learning and converge as early as possible, we divide learning into two steps. The first is rapid learning. In the phase, we choose a learning rate which is bigger than When receiving stable weight vectors, we transfer the first phase to the second phase, slow learning. We set learning rate to 0.05 in the second phase. In order to avoid weight of neurons in the competitive layer be not able to converge during learning process, we select a prototype vector from each subclass of spare s and legitimate s as an initial weight vector respectively. 4. Experiments and result This project makes use of corpus from which is open available source. We select 1000 pieces s randomly from the corpus, including 580 spare s, 420 legitimate s. The corpus consists of all English s whose attachments, html tags and headers except the subject line have been stripped off. 37

5 Anti-spare filter performance is often measured in terms of spam precision (SP) and sparn recall (SR): SP = nsp n'p m-~'p "... pan + nleg,t.,pam (7) n,~,_.,~, are the numbers of spam s that the filter classified as spare s, nt,~,~,p,= are the numbers of legitimate s that the filter classified as spare s mistakenly. SR = Bspam"~'sPara N'v m (8) Nspm are the total numbers of spam s spare recall measures the percentage of spare that the filter manages to block (intuitively its effectiveness), while spare precision measures the degree to which the blocked s are indeed spare (the filter's safety), which is more important factor in the performance evaluation. On our intuitiveness, it is difficult to compare the performance of different filters using spare recall and precision: each filter (or filter configuration) yields a pair of spare recall and precision results; without a single unifying measure. Therefore, we introduce a criterion F1, which incorporates spam precision and spare recall. It is defined: SP x SR x 2 FI= se + SR (9) In the experiments, first of all, we compare performance of the LVQ network in different training times on both open set and close set, shown in table 1. Then, we compare performances of LVQ-based approach, Bayes-based as well as BP neural network-based, shown in table 2 Table 1. Result of Test I Training times SP(%) Open set Close set SR(%) SP(%) SR(%) Table 2. Result of Test 2 SP(%) SR(%) Fl(%) N~ffve Bayes ANN--BP ANN--LVQ When the numbers of training of our LVQ network reache 500, SP and SR of the filter are not very ideal, after coming to 1000, the performances improve apparently, when reaching 1500, the performances only have a little improvement than before. In open set, SP is 98.97%, SR is 93.5%, in close set, SP is 99.51%, SR is 96.86% In table 2, we list the performances of three algorithms, both algorithms based on neural network, which have slight improvement in SP whereas improve SR apparently, are superior to Bayes-based approach. In two neural networks, LVQ-based is better than BP-based in both SP and SR. In the terms of F 1, Na'fve Bayes, ANN-BP, ANN-LVQ increase in sequence. 38

6 5. Conclusion ANN-LVQ approach further classifies spare into several subclasses according to spam s' content in order that s are easy to identify. Then it combines the subclasses into complex classes by LVQ neural network so as to identify spare s. It is proved by experiments The numbers of neural network training affect the performance of the filter, when the numbers of training are not enough, the performance of the filter is not ideal, when the numbers reache 1500, the performance gets to stability. Both neural network-based algorithms are usually better than that based on Bayes. Because neural networks take account of relationship between each feature words on the whole, and yet Bayse-based algorithm simply thinks that feature words are independent. LVQ-based method outperforms that based on BP, because we classify spam s into several subclasses in content so that the feature words of each subclass of spare is more related and closer as well as characteristics of each subclass of spare s are easier to identify. References [1] CNNIC, the 13th China Intemet development status report, 2004, 1 [2] IResearch Inc. China Anti-Spare Market Research Report, 2003, 11 [3] William W. Cohen. Learning rules that classify . In proceedings of the 1996 AAAI Spring symposium in information access, 1996 [4] Sahami, M, S. Dumais, et al. A Bayesian Approach to Filtering Junk . Learing for Text Categorization -Papers from the AAAI Workshop, Madison Wisconsin [5] X. Carreras and L. Mrquez. Boosting trees for anti-spare filtering. In Proceedings of RANLP-01, Jth International Conference on Recent Advances in Natural Language Processing, Tzigov Chark, BG, 2001 [6] Duhong chen, Tongjie et al. Spare Filter Using Naive Bayesian, Decision Tree, Neural Network and AdaBoost, [7] James Clark, Ima Koprinska, Josiah Poon, A neural network based approach to automated classification, Proceedings of the IEEE/WlC international conference on web intelligence. [8] Salton G, Wong A, Yang C S. A vector space model for automatic indexing. Communication s of the ACM, [9] Salton G. Introduction to modem information retrieval. New York McGraw-Hill Book company [10] Kenneth Ward Church and Tatrick Hanks. Workd association norms, mutual information and lexicography. In proceedings ofacl27, Wancouver, Canada, [11] China anti-spare market research report in 2004, IResearch Inc [12] Martin T.Hagan, Howard B.Demuth, Nark H. Beale, Neural network design, China Machine Press, 2002,8 39

A Proposed Algorithm for Spam Filtering Emails by Hash Table Approach

A Proposed Algorithm for Spam Filtering Emails by Hash Table Approach International Research Journal of Applied and Basic Sciences 2013 Available online at www.irjabs.com ISSN 2251-838X / Vol, 4 (9): 2436-2441 Science Explorer Publications A Proposed Algorithm for Spam Filtering

More information

A Personalized Spam Filtering Approach Utilizing Two Separately Trained Filters

A Personalized Spam Filtering Approach Utilizing Two Separately Trained Filters 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology A Personalized Spam Filtering Approach Utilizing Two Separately Trained Filters Wei-Lun Teng, Wei-Chung Teng

More information

CAS-ICT at TREC 2005 SPAM Track: Using Non-Textual Information to Improve Spam Filtering Performance

CAS-ICT at TREC 2005 SPAM Track: Using Non-Textual Information to Improve Spam Filtering Performance CAS-ICT at TREC 2005 SPAM Track: Using Non-Textual Information to Improve Spam Filtering Performance Shen Wang, Bin Wang and Hao Lang, Xueqi Cheng Institute of Computing Technology, Chinese Academy of

More information

Bayesian Spam Filtering

Bayesian Spam Filtering Bayesian Spam Filtering Ahmed Obied Department of Computer Science University of Calgary amaobied@ucalgary.ca http://www.cpsc.ucalgary.ca/~amaobied Abstract. With the enormous amount of spam messages propagating

More information

Filtering Junk Mail with A Maximum Entropy Model

Filtering Junk Mail with A Maximum Entropy Model Filtering Junk Mail with A Maximum Entropy Model ZHANG Le and YAO Tian-shun Institute of Computer Software & Theory. School of Information Science & Engineering, Northeastern University Shenyang, 110004

More information

An Efficient Two-phase Spam Filtering Method Based on E-mails Categorization

An Efficient Two-phase Spam Filtering Method Based on E-mails Categorization International Journal of Network Security, Vol.9, No., PP.34 43, July 29 34 An Efficient Two-phase Spam Filtering Method Based on E-mails Categorization Jyh-Jian Sheu Department of Information Management,

More information

A Content based Spam Filtering Using Optical Back Propagation Technique

A Content based Spam Filtering Using Optical Back Propagation Technique A Content based Spam Filtering Using Optical Back Propagation Technique Sarab M. Hameed 1, Noor Alhuda J. Mohammed 2 Department of Computer Science, College of Science, University of Baghdad - Iraq ABSTRACT

More information

Social Media Mining. Data Mining Essentials

Social Media Mining. Data Mining Essentials Introduction Data production rate has been increased dramatically (Big Data) and we are able store much more data than before E.g., purchase data, social media data, mobile phone data Businesses and customers

More information

Representation of Electronic Mail Filtering Profiles: A User Study

Representation of Electronic Mail Filtering Profiles: A User Study Representation of Electronic Mail Filtering Profiles: A User Study Michael J. Pazzani Department of Information and Computer Science University of California, Irvine Irvine, CA 92697 +1 949 824 5888 pazzani@ics.uci.edu

More information

Anti-Spam Filter Based on Naïve Bayes, SVM, and KNN model

Anti-Spam Filter Based on Naïve Bayes, SVM, and KNN model AI TERM PROJECT GROUP 14 1 Anti-Spam Filter Based on,, and model Yun-Nung Chen, Che-An Lu, Chao-Yu Huang Abstract spam email filters are a well-known and powerful type of filters. We construct different

More information

WEB PAGE CATEGORISATION BASED ON NEURONS

WEB PAGE CATEGORISATION BASED ON NEURONS WEB PAGE CATEGORISATION BASED ON NEURONS Shikha Batra Abstract: Contemporary web is comprised of trillions of pages and everyday tremendous amount of requests are made to put more web pages on the WWW.

More information

agoweder@yahoo.com ** The High Institute of Zahra for Comperhensive Professions, Zahra-Libya

agoweder@yahoo.com ** The High Institute of Zahra for Comperhensive Professions, Zahra-Libya AN ANTI-SPAM SYSTEM USING ARTIFICIAL NEURAL NETWORKS AND GENETIC ALGORITHMS ABDUELBASET M. GOWEDER *, TARIK RASHED **, ALI S. ELBEKAIE ***, and HUSIEN A. ALHAMMI **** * The High Institute of Surman for

More information

Detecting E-mail Spam Using Spam Word Associations

Detecting E-mail Spam Using Spam Word Associations Detecting E-mail Spam Using Spam Word Associations N.S. Kumar 1, D.P. Rana 2, R.G.Mehta 3 Sardar Vallabhbhai National Institute of Technology, Surat, India 1 p10co977@coed.svnit.ac.in 2 dpr@coed.svnit.ac.in

More information

Three types of messages: A, B, C. Assume A is the oldest type, and C is the most recent type.

Three types of messages: A, B, C. Assume A is the oldest type, and C is the most recent type. Chronological Sampling for Email Filtering Ching-Lung Fu 2, Daniel Silver 1, and James Blustein 2 1 Acadia University, Wolfville, Nova Scotia, Canada 2 Dalhousie University, Halifax, Nova Scotia, Canada

More information

An Imbalanced Spam Mail Filtering Method

An Imbalanced Spam Mail Filtering Method , pp. 119-126 http://dx.doi.org/10.14257/ijmue.2015.10.3.12 An Imbalanced Spam Mail Filtering Method Zhiqiang Ma, Rui Yan, Donghong Yuan and Limin Liu (College of Information Engineering, Inner Mongolia

More information

Three-Way Decisions Solution to Filter Spam Email: An Empirical Study

Three-Way Decisions Solution to Filter Spam Email: An Empirical Study Three-Way Decisions Solution to Filter Spam Email: An Empirical Study Xiuyi Jia 1,4, Kan Zheng 2,WeiweiLi 3, Tingting Liu 2, and Lin Shang 4 1 School of Computer Science and Technology, Nanjing University

More information

Spam Detection System Combining Cellular Automata and Naive Bayes Classifier

Spam Detection System Combining Cellular Automata and Naive Bayes Classifier Spam Detection System Combining Cellular Automata and Naive Bayes Classifier F. Barigou*, N. Barigou**, B. Atmani*** Computer Science Department, Faculty of Sciences, University of Oran BP 1524, El M Naouer,

More information

A MACHINE LEARNING APPROACH TO SERVER-SIDE ANTI-SPAM E-MAIL FILTERING 1 2

A MACHINE LEARNING APPROACH TO SERVER-SIDE ANTI-SPAM E-MAIL FILTERING 1 2 UDC 004.75 A MACHINE LEARNING APPROACH TO SERVER-SIDE ANTI-SPAM E-MAIL FILTERING 1 2 I. Mashechkin, M. Petrovskiy, A. Rozinkin, S. Gerasimov Computer Science Department, Lomonosov Moscow State University,

More information

Active Learning SVM for Blogs recommendation

Active Learning SVM for Blogs recommendation Active Learning SVM for Blogs recommendation Xin Guan Computer Science, George Mason University Ⅰ.Introduction In the DH Now website, they try to review a big amount of blogs and articles and find the

More information

Impact of Feature Selection Technique on Email Classification

Impact of Feature Selection Technique on Email Classification Impact of Feature Selection Technique on Email Classification Aakanksha Sharaff, Naresh Kumar Nagwani, and Kunal Swami Abstract Being one of the most powerful and fastest way of communication, the popularity

More information

Email Spam Detection A Machine Learning Approach

Email Spam Detection A Machine Learning Approach Email Spam Detection A Machine Learning Approach Ge Song, Lauren Steimle ABSTRACT Machine learning is a branch of artificial intelligence concerned with the creation and study of systems that can learn

More information

IMPROVING SPAM EMAIL FILTERING EFFICIENCY USING BAYESIAN BACKWARD APPROACH PROJECT

IMPROVING SPAM EMAIL FILTERING EFFICIENCY USING BAYESIAN BACKWARD APPROACH PROJECT IMPROVING SPAM EMAIL FILTERING EFFICIENCY USING BAYESIAN BACKWARD APPROACH PROJECT M.SHESHIKALA Assistant Professor, SREC Engineering College,Warangal Email: marthakala08@gmail.com, Abstract- Unethical

More information

Abstract. Find out if your mortgage rate is too high, NOW. Free Search

Abstract. Find out if your mortgage rate is too high, NOW. Free Search Statistics and The War on Spam David Madigan Rutgers University Abstract Text categorization algorithms assign texts to predefined categories. The study of such algorithms has a rich history dating back

More information

MODULE 15 Clustering Large Datasets LESSON 34

MODULE 15 Clustering Large Datasets LESSON 34 MODULE 15 Clustering Large Datasets LESSON 34 Incremental Clustering Keywords: Single Database Scan, Leader, BIRCH, Tree 1 Clustering Large Datasets Pattern matrix It is convenient to view the input data

More information

Email Spam Detection Using Customized SimHash Function

Email Spam Detection Using Customized SimHash Function International Journal of Research Studies in Computer Science and Engineering (IJRSCSE) Volume 1, Issue 8, December 2014, PP 35-40 ISSN 2349-4840 (Print) & ISSN 2349-4859 (Online) www.arcjournals.org Email

More information

Chapter 2 The Research on Fault Diagnosis of Building Electrical System Based on RBF Neural Network

Chapter 2 The Research on Fault Diagnosis of Building Electrical System Based on RBF Neural Network Chapter 2 The Research on Fault Diagnosis of Building Electrical System Based on RBF Neural Network Qian Wu, Yahui Wang, Long Zhang and Li Shen Abstract Building electrical system fault diagnosis is the

More information

Naive Bayes Spam Filtering Using Word-Position-Based Attributes

Naive Bayes Spam Filtering Using Word-Position-Based Attributes Naive Bayes Spam Filtering Using Word-Position-Based Attributes Johan Hovold Department of Computer Science Lund University Box 118, 221 00 Lund, Sweden johan.hovold.363@student.lu.se Abstract This paper

More information

An Efficient Spam Filtering Techniques for Email Account

An Efficient Spam Filtering Techniques for Email Account American Journal of Engineering Research (AJER) e-issn : 2320-0847 p-issn : 2320-0936 Volume-02, Issue-10, pp-63-73 www.ajer.org Research Paper Open Access An Efficient Spam Filtering Techniques for Email

More information

Spam Filtering using Naïve Bayesian Classification

Spam Filtering using Naïve Bayesian Classification Spam Filtering using Naïve Bayesian Classification Presented by: Samer Younes Outline What is spam anyway? Some statistics Why is Spam a Problem Major Techniques for Classifying Spam Transport Level Filtering

More information

WE DEFINE spam as an e-mail message that is unwanted basically

WE DEFINE spam as an e-mail message that is unwanted basically 1048 IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 10, NO. 5, SEPTEMBER 1999 Support Vector Machines for Spam Categorization Harris Drucker, Senior Member, IEEE, Donghui Wu, Student Member, IEEE, and Vladimir

More information

Learning to classify e-mail

Learning to classify e-mail Information Sciences 177 (2007) 2167 2187 www.elsevier.com/locate/ins Learning to classify e-mail Irena Koprinska *, Josiah Poon, James Clark, Jason Chan School of Information Technologies, The University

More information

Simple Language Models for Spam Detection

Simple Language Models for Spam Detection Simple Language Models for Spam Detection Egidio Terra Faculty of Informatics PUC/RS - Brazil Abstract For this year s Spam track we used classifiers based on language models. These models are used to

More information

Naïve Bayesian Anti-spam Filtering Technique for Malay Language

Naïve Bayesian Anti-spam Filtering Technique for Malay Language Naïve Bayesian Anti-spam Filtering Technique for Malay Language Thamarai Subramaniam 1, Hamid A. Jalab 2, Alaa Y. Taqa 3 1,2 Computer System and Technology Department, Faulty of Computer Science and Information

More information

Image Content-Based Email Spam Image Filtering

Image Content-Based Email Spam Image Filtering Image Content-Based Email Spam Image Filtering Jianyi Wang and Kazuki Katagishi Abstract With the population of Internet around the world, email has become one of the main methods of communication among

More information

Artificial Neural Network, Decision Tree and Statistical Techniques Applied for Designing and Developing E-mail Classifier

Artificial Neural Network, Decision Tree and Statistical Techniques Applied for Designing and Developing E-mail Classifier International Journal of Recent Technology and Engineering (IJRTE) ISSN: 2277-3878, Volume-1, Issue-6, January 2013 Artificial Neural Network, Decision Tree and Statistical Techniques Applied for Designing

More information

Detecting Spam Bots in Online Social Networking Sites: A Machine Learning Approach

Detecting Spam Bots in Online Social Networking Sites: A Machine Learning Approach Detecting Spam Bots in Online Social Networking Sites: A Machine Learning Approach Alex Hai Wang College of Information Sciences and Technology, The Pennsylvania State University, Dunmore, PA 18512, USA

More information

Large-Scale Data Sets Clustering Based on MapReduce and Hadoop

Large-Scale Data Sets Clustering Based on MapReduce and Hadoop Journal of Computational Information Systems 7: 16 (2011) 5956-5963 Available at http://www.jofcis.com Large-Scale Data Sets Clustering Based on MapReduce and Hadoop Ping ZHOU, Jingsheng LEI, Wenjun YE

More information

Construction Algorithms for Index Model Based on Web Page Classification

Construction Algorithms for Index Model Based on Web Page Classification Journal of Computational Information Systems 10: 2 (2014) 655 664 Available at http://www.jofcis.com Construction Algorithms for Index Model Based on Web Page Classification Yangjie ZHANG 1,2,, Chungang

More information

SpamNet Spam Detection Using PCA and Neural Networks

SpamNet Spam Detection Using PCA and Neural Networks SpamNet Spam Detection Using PCA and Neural Networks Abhimanyu Lad B.Tech. (I.T.) 4 th year student Indian Institute of Information Technology, Allahabad Deoghat, Jhalwa, Allahabad, India abhimanyulad@iiita.ac.in

More information

Machine Learning using MapReduce

Machine Learning using MapReduce Machine Learning using MapReduce What is Machine Learning Machine learning is a subfield of artificial intelligence concerned with techniques that allow computers to improve their outputs based on previous

More information

Web Document Clustering

Web Document Clustering Web Document Clustering Lab Project based on the MDL clustering suite http://www.cs.ccsu.edu/~markov/mdlclustering/ Zdravko Markov Computer Science Department Central Connecticut State University New Britain,

More information

A Three-Way Decision Approach to Email Spam Filtering

A Three-Way Decision Approach to Email Spam Filtering A Three-Way Decision Approach to Email Spam Filtering Bing Zhou, Yiyu Yao, and Jigang Luo Department of Computer Science, University of Regina Regina, Saskatchewan, Canada S4S 0A2 {zhou200b,yyao,luo226}@cs.uregina.ca

More information

Stabilization by Conceptual Duplication in Adaptive Resonance Theory

Stabilization by Conceptual Duplication in Adaptive Resonance Theory Stabilization by Conceptual Duplication in Adaptive Resonance Theory Louis Massey Royal Military College of Canada Department of Mathematics and Computer Science PO Box 17000 Station Forces Kingston, Ontario,

More information

A FUZZY BASED APPROACH TO TEXT MINING AND DOCUMENT CLUSTERING

A FUZZY BASED APPROACH TO TEXT MINING AND DOCUMENT CLUSTERING A FUZZY BASED APPROACH TO TEXT MINING AND DOCUMENT CLUSTERING Sumit Goswami 1 and Mayank Singh Shishodia 2 1 Indian Institute of Technology-Kharagpur, Kharagpur, India sumit_13@yahoo.com 2 School of Computer

More information

Machine Learning Final Project Spam Email Filtering

Machine Learning Final Project Spam Email Filtering Machine Learning Final Project Spam Email Filtering March 2013 Shahar Yifrah Guy Lev Table of Content 1. OVERVIEW... 3 2. DATASET... 3 2.1 SOURCE... 3 2.2 CREATION OF TRAINING AND TEST SETS... 4 2.3 FEATURE

More information

Automatic Web Page Classification

Automatic Web Page Classification Automatic Web Page Classification Yasser Ganjisaffar 84802416 yganjisa@uci.edu 1 Introduction To facilitate user browsing of Web, some websites such as Yahoo! (http://dir.yahoo.com) and Open Directory

More information

INTERNATIONAL JOURNAL OF ADVANCES IN COMPUTING AND INFORMATION TECHNOLOGY An International online open access peer reviewed journal

INTERNATIONAL JOURNAL OF ADVANCES IN COMPUTING AND INFORMATION TECHNOLOGY An International online open access peer reviewed journal INTERNATIONAL JOURNAL OF ADVANCES IN COMPUTING AND INFORMATION TECHNOLOGY An International online open access peer reviewed journal Research Article ISSN 2277 9140 ABSTRACT Web page categorization based

More information

Clustering Technique in Data Mining for Text Documents

Clustering Technique in Data Mining for Text Documents Clustering Technique in Data Mining for Text Documents Ms.J.Sathya Priya Assistant Professor Dept Of Information Technology. Velammal Engineering College. Chennai. Ms.S.Priyadharshini Assistant Professor

More information

Question 2 Naïve Bayes (16 points)

Question 2 Naïve Bayes (16 points) Question 2 Naïve Bayes (16 points) About 2/3 of your email is spam so you downloaded an open source spam filter based on word occurrences that uses the Naive Bayes classifier. Assume you collected the

More information

A Two-Pass Statistical Approach for Automatic Personalized Spam Filtering

A Two-Pass Statistical Approach for Automatic Personalized Spam Filtering A Two-Pass Statistical Approach for Automatic Personalized Spam Filtering Khurum Nazir Junejo, Mirza Muhammad Yousaf, and Asim Karim Dept. of Computer Science, Lahore University of Management Sciences

More information

Design call center management system of e-commerce based on BP neural network and multifractal

Design call center management system of e-commerce based on BP neural network and multifractal Available online www.jocpr.com Journal of Chemical and Pharmaceutical Research, 2014, 6(6):951-956 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 Design call center management system of e-commerce

More information

6367(Print), ISSN 0976 6375(Online) & TECHNOLOGY Volume 4, Issue 1, (IJCET) January- February (2013), IAEME

6367(Print), ISSN 0976 6375(Online) & TECHNOLOGY Volume 4, Issue 1, (IJCET) January- February (2013), IAEME INTERNATIONAL International Journal of Computer JOURNAL Engineering OF COMPUTER and Technology ENGINEERING (IJCET), ISSN 0976-6367(Print), ISSN 0976 6375(Online) & TECHNOLOGY Volume 4, Issue 1, (IJCET)

More information

MACHINE LEARNING. Introduction. Alessandro Moschitti

MACHINE LEARNING. Introduction. Alessandro Moschitti MACHINE LEARNING Introduction Alessandro Moschitti Department of Computer Science and Information Engineering University of Trento Email: moschitti@disi.unitn.it Course Schedule Lectures Tuesday, 14:00-16:00

More information

PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 4: LINEAR MODELS FOR CLASSIFICATION

PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 4: LINEAR MODELS FOR CLASSIFICATION PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 4: LINEAR MODELS FOR CLASSIFICATION Introduction In the previous chapter, we explored a class of regression models having particularly simple analytical

More information

Lan, Mingjun and Zhou, Wanlei 2005, Spam filtering based on preference ranking, in Fifth International Conference on Computer and Information

Lan, Mingjun and Zhou, Wanlei 2005, Spam filtering based on preference ranking, in Fifth International Conference on Computer and Information Lan, Mingjun and Zhou, Wanlei 2005, Spam filtering based on preference ranking, in Fifth International Conference on Computer and Information Technology : CIT 2005 : proceedings : 21-23 September, 2005,

More information

E-mail Spam Classification With Artificial Neural Network and Negative Selection Algorithm

E-mail Spam Classification With Artificial Neural Network and Negative Selection Algorithm E-mail Spam Classification With Artificial Neural Network and Negative Selection Algorithm Ismaila Idris Dept of Cyber Security Science, Federal University of Technology, Minna, Nigeria. Idris.ismaila95@gmail.com

More information

Chapter 6. The stacking ensemble approach

Chapter 6. The stacking ensemble approach 82 This chapter proposes the stacking ensemble approach for combining different data mining classifiers to get better performance. Other combination techniques like voting, bagging etc are also described

More information

EM Clustering Approach for Multi-Dimensional Analysis of Big Data Set

EM Clustering Approach for Multi-Dimensional Analysis of Big Data Set EM Clustering Approach for Multi-Dimensional Analysis of Big Data Set Amhmed A. Bhih School of Electrical and Electronic Engineering Princy Johnson School of Electrical and Electronic Engineering Martin

More information

SPECIAL PERTURBATIONS UNCORRELATED TRACK PROCESSING

SPECIAL PERTURBATIONS UNCORRELATED TRACK PROCESSING AAS 07-228 SPECIAL PERTURBATIONS UNCORRELATED TRACK PROCESSING INTRODUCTION James G. Miller * Two historical uncorrelated track (UCT) processing approaches have been employed using general perturbations

More information

T-61.3050 : Email Classification as Spam or Ham using Naive Bayes Classifier. Santosh Tirunagari : 245577

T-61.3050 : Email Classification as Spam or Ham using Naive Bayes Classifier. Santosh Tirunagari : 245577 T-61.3050 : Email Classification as Spam or Ham using Naive Bayes Classifier Santosh Tirunagari : 245577 January 20, 2011 Abstract This term project gives a solution how to classify an email as spam or

More information

Categorical Data Visualization and Clustering Using Subjective Factors

Categorical Data Visualization and Clustering Using Subjective Factors Categorical Data Visualization and Clustering Using Subjective Factors Chia-Hui Chang and Zhi-Kai Ding Department of Computer Science and Information Engineering, National Central University, Chung-Li,

More information

Predict Influencers in the Social Network

Predict Influencers in the Social Network Predict Influencers in the Social Network Ruishan Liu, Yang Zhao and Liuyu Zhou Email: rliu2, yzhao2, lyzhou@stanford.edu Department of Electrical Engineering, Stanford University Abstract Given two persons

More information

Clustering. Danilo Croce Web Mining & Retrieval a.a. 2015/201 16/03/2016

Clustering. Danilo Croce Web Mining & Retrieval a.a. 2015/201 16/03/2016 Clustering Danilo Croce Web Mining & Retrieval a.a. 2015/201 16/03/2016 1 Supervised learning vs. unsupervised learning Supervised learning: discover patterns in the data that relate data attributes with

More information

Single-Class Learning for Spam Filtering: An Ensemble Approach

Single-Class Learning for Spam Filtering: An Ensemble Approach Single-Class Learning for Spam Filtering: An Ensemble Approach Tsang-Hsiang Cheng Department of Business Administration Southern Taiwan University of Technology Tainan, Taiwan, R.O.C. Chih-Ping Wei Institute

More information

The Enron Corpus: A New Dataset for Email Classification Research

The Enron Corpus: A New Dataset for Email Classification Research The Enron Corpus: A New Dataset for Email Classification Research Bryan Klimt and Yiming Yang Language Technologies Institute Carnegie Mellon University Pittsburgh, PA 15213-8213, USA {bklimt,yiming}@cs.cmu.edu

More information

Machine Learning CS 6830. Lecture 01. Razvan C. Bunescu School of Electrical Engineering and Computer Science bunescu@ohio.edu

Machine Learning CS 6830. Lecture 01. Razvan C. Bunescu School of Electrical Engineering and Computer Science bunescu@ohio.edu Machine Learning CS 6830 Razvan C. Bunescu School of Electrical Engineering and Computer Science bunescu@ohio.edu What is Learning? Merriam-Webster: learn = to acquire knowledge, understanding, or skill

More information

Statistical Validation and Data Analytics in ediscovery. Jesse Kornblum

Statistical Validation and Data Analytics in ediscovery. Jesse Kornblum Statistical Validation and Data Analytics in ediscovery Jesse Kornblum Administrivia Silence your mobile Interactive talk Please ask questions 2 Outline Introduction Big Questions What Makes Things Similar?

More information

Email Classification Using Data Reduction Method

Email Classification Using Data Reduction Method Email Classification Using Data Reduction Method Rafiqul Islam and Yang Xiang, member IEEE School of Information Technology Deakin University, Burwood 3125, Victoria, Australia Abstract Classifying user

More information

Open Access Research on Application of Neural Network in Computer Network Security Evaluation. Shujuan Jin *

Open Access Research on Application of Neural Network in Computer Network Security Evaluation. Shujuan Jin * Send Orders for Reprints to reprints@benthamscience.ae 766 The Open Electrical & Electronic Engineering Journal, 2014, 8, 766-771 Open Access Research on Application of Neural Network in Computer Network

More information

Network Machine Learning Research Group. Intended status: Informational October 19, 2015 Expires: April 21, 2016

Network Machine Learning Research Group. Intended status: Informational October 19, 2015 Expires: April 21, 2016 Network Machine Learning Research Group S. Jiang Internet-Draft Huawei Technologies Co., Ltd Intended status: Informational October 19, 2015 Expires: April 21, 2016 Abstract Network Machine Learning draft-jiang-nmlrg-network-machine-learning-00

More information

VCU-TSA at Semeval-2016 Task 4: Sentiment Analysis in Twitter

VCU-TSA at Semeval-2016 Task 4: Sentiment Analysis in Twitter VCU-TSA at Semeval-2016 Task 4: Sentiment Analysis in Twitter Gerard Briones and Kasun Amarasinghe and Bridget T. McInnes, PhD. Department of Computer Science Virginia Commonwealth University Richmond,

More information

A Composite Intelligent Method for Spam Filtering

A Composite Intelligent Method for Spam Filtering , pp.67-76 http://dx.doi.org/10.14257/ijsia.2014.8.4.07 A Composite Intelligent Method for Spam Filtering Jun Liu 1*, Shuyu Chen 2, Kai Liu 1 and ong Zhou 1 1 College of Computer Science, Chongqing University,

More information

E-commerce Transaction Anomaly Classification

E-commerce Transaction Anomaly Classification E-commerce Transaction Anomaly Classification Minyong Lee minyong@stanford.edu Seunghee Ham sham12@stanford.edu Qiyi Jiang qjiang@stanford.edu I. INTRODUCTION Due to the increasing popularity of e-commerce

More information

Content-Based Recommendation

Content-Based Recommendation Content-Based Recommendation Content-based? Item descriptions to identify items that are of particular interest to the user Example Example Comparing with Noncontent based Items User-based CF Searches

More information

A Game Theoretical Framework for Adversarial Learning

A Game Theoretical Framework for Adversarial Learning A Game Theoretical Framework for Adversarial Learning Murat Kantarcioglu University of Texas at Dallas Richardson, TX 75083, USA muratk@utdallas Chris Clifton Purdue University West Lafayette, IN 47907,

More information

Towards better accuracy for Spam predictions

Towards better accuracy for Spam predictions Towards better accuracy for Spam predictions Chengyan Zhao Department of Computer Science University of Toronto Toronto, Ontario, Canada M5S 2E4 czhao@cs.toronto.edu Abstract Spam identification is crucial

More information

Feature Subset Selection in E-mail Spam Detection

Feature Subset Selection in E-mail Spam Detection Feature Subset Selection in E-mail Spam Detection Amir Rajabi Behjat, Universiti Technology MARA, Malaysia IT Security for the Next Generation Asia Pacific & MEA Cup, Hong Kong 14-16 March, 2012 Feature

More information

Web Page Categorization based on Document Structure

Web Page Categorization based on Document Structure 1 Web Page Categorization based on Document Structure Arul Prakash Asirvatham arul@gdit.iiit.net Kranthi Kumar. Ravi kranthi@gdit.iiit.net Centre for Visual Information Technology International Institute

More information

The Data Mining Process

The Data Mining Process Sequence for Determining Necessary Data. Wrong: Catalog everything you have, and decide what data is important. Right: Work backward from the solution, define the problem explicitly, and map out the data

More information

Machine Learning for Naive Bayesian Spam Filter Tokenization

Machine Learning for Naive Bayesian Spam Filter Tokenization Machine Learning for Naive Bayesian Spam Filter Tokenization Michael Bevilacqua-Linn December 20, 2003 Abstract Background Traditional client level spam filters rely on rule based heuristics. While these

More information

Hoodwinking Spam Email Filters

Hoodwinking Spam Email Filters Proceedings of the 2007 WSEAS International Conference on Computer Engineering and Applications, Gold Coast, Australia, January 17-19, 2007 533 Hoodwinking Spam Email Filters WANLI MA, DAT TRAN, DHARMENDRA

More information

Combining SVM classifiers for email anti-spam filtering

Combining SVM classifiers for email anti-spam filtering Combining SVM classifiers for email anti-spam filtering Ángela Blanco Manuel Martín-Merino Abstract Spam, also known as Unsolicited Commercial Email (UCE) is becoming a nightmare for Internet users and

More information

Comparative Study of Features Space Reduction Techniques for Spam Detection

Comparative Study of Features Space Reduction Techniques for Spam Detection Comparative Study of Features Space Reduction Techniques for Spam Detection By Nouman Azam 1242 (MS-5) Supervised by Dr. Amir Hanif Dar Thesis committee Brig. Dr Muhammad Younas Javed Dr. Azad A Saddiqui

More information

A Partially Supervised Metric Multidimensional Scaling Algorithm for Textual Data Visualization

A Partially Supervised Metric Multidimensional Scaling Algorithm for Textual Data Visualization A Partially Supervised Metric Multidimensional Scaling Algorithm for Textual Data Visualization Ángela Blanco Universidad Pontificia de Salamanca ablancogo@upsa.es Spain Manuel Martín-Merino Universidad

More information

Differential Voting in Case Based Spam Filtering

Differential Voting in Case Based Spam Filtering Differential Voting in Case Based Spam Filtering Deepak P, Delip Rao, Deepak Khemani Department of Computer Science and Engineering Indian Institute of Technology Madras, India deepakswallet@gmail.com,

More information

Less naive Bayes spam detection

Less naive Bayes spam detection Less naive Bayes spam detection Hongming Yang Eindhoven University of Technology Dept. EE, Rm PT 3.27, P.O.Box 53, 5600MB Eindhoven The Netherlands. E-mail:h.m.yang@tue.nl also CoSiNe Connectivity Systems

More information

FUZZY CLUSTERING ANALYSIS OF DATA MINING: APPLICATION TO AN ACCIDENT MINING SYSTEM

FUZZY CLUSTERING ANALYSIS OF DATA MINING: APPLICATION TO AN ACCIDENT MINING SYSTEM International Journal of Innovative Computing, Information and Control ICIC International c 0 ISSN 34-48 Volume 8, Number 8, August 0 pp. 4 FUZZY CLUSTERING ANALYSIS OF DATA MINING: APPLICATION TO AN ACCIDENT

More information

Improving spam mail filtering using classification algorithms with discretization Filter

Improving spam mail filtering using classification algorithms with discretization Filter International Association of Scientific Innovation and Research (IASIR) (An Association Unifying the Sciences, Engineering, and Applied Research) International Journal of Emerging Technologies in Computational

More information

MASTER THESIS IN SOFTWARE CONSTRUCTION AUTOMATIC E-MAIL CATEGORIZATION

MASTER THESIS IN SOFTWARE CONSTRUCTION AUTOMATIC E-MAIL CATEGORIZATION MASTER THESIS IN SOFTWARE CONSTRUCTION AUTOMATIC E-MAIL CATEGORIZATION Author: Damian Bukszynski Dbuksz11@student.aau.dk Supervisor: Dr. Daniel Ortiz-Arroyo do@cs.aaue.dk October 1, 2013 SUMMARY Nowadays

More information

LVQ Plug-In Algorithm for SQL Server

LVQ Plug-In Algorithm for SQL Server LVQ Plug-In Algorithm for SQL Server Licínia Pedro Monteiro Instituto Superior Técnico licinia.monteiro@tagus.ist.utl.pt I. Executive Summary In this Resume we describe a new functionality implemented

More information

A Novel Spam Email Detection System Based on Negative Selection

A Novel Spam Email Detection System Based on Negative Selection 2009 Fourth International Conference on Computer Sciences and Convergence Information Technology A Novel Spam Email Detection System Based on Negative Selection Wanli Ma, Dat Tran, and Dharmendra Sharma

More information

Neural Networks and Back Propagation Algorithm

Neural Networks and Back Propagation Algorithm Neural Networks and Back Propagation Algorithm Mirza Cilimkovic Institute of Technology Blanchardstown Blanchardstown Road North Dublin 15 Ireland mirzac@gmail.com Abstract Neural Networks (NN) are important

More information

Term extraction for user profiling: evaluation by the user

Term extraction for user profiling: evaluation by the user Term extraction for user profiling: evaluation by the user Suzan Verberne 1, Maya Sappelli 1,2, Wessel Kraaij 1,2 1 Institute for Computing and Information Sciences, Radboud University Nijmegen 2 TNO,

More information

An Approach to Detect Spam Emails by Using Majority Voting

An Approach to Detect Spam Emails by Using Majority Voting An Approach to Detect Spam Emails by Using Majority Voting Roohi Hussain Department of Computer Engineering, National University of Science and Technology, H-12 Islamabad, Pakistan Usman Qamar Faculty,

More information

Spam E-mail Detection by Random Forests Algorithm

Spam E-mail Detection by Random Forests Algorithm Spam E-mail Detection by Random Forests Algorithm 1 Bhagyashri U. Gaikwad, 2 P. P. Halkarnikar 1 M. Tech Student, Computer Science and Technology, Department of Technology, Shivaji University, Kolhapur,

More information

Behavior Analysis of SVM Based Spam Filtering Using Various Kernel Functions and Data Representations

Behavior Analysis of SVM Based Spam Filtering Using Various Kernel Functions and Data Representations ISSN: 2278-181 Vol. 2 Issue 9, September - 213 Behavior Analysis of SVM Based Spam Filtering Using Various Kernel Functions and Data Representations Author :Sushama Chouhan Author Affiliation: MTech Scholar

More information

SVM-Based Spam Filter with Active and Online Learning

SVM-Based Spam Filter with Active and Online Learning SVM-Based Spam Filter with Active and Online Learning Qiang Wang Yi Guan Xiaolong Wang School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China Email:{qwang, guanyi,

More information

Mobile Phone APP Software Browsing Behavior using Clustering Analysis

Mobile Phone APP Software Browsing Behavior using Clustering Analysis Proceedings of the 2014 International Conference on Industrial Engineering and Operations Management Bali, Indonesia, January 7 9, 2014 Mobile Phone APP Software Browsing Behavior using Clustering Analysis

More information

SURVEY PAPER ON INTELLIGENT SYSTEM FOR TEXT AND IMAGE SPAM FILTERING Amol H. Malge 1, Dr. S. M. Chaware 2

SURVEY PAPER ON INTELLIGENT SYSTEM FOR TEXT AND IMAGE SPAM FILTERING Amol H. Malge 1, Dr. S. M. Chaware 2 International Journal of Computer Engineering and Applications, Volume IX, Issue I, January 15 SURVEY PAPER ON INTELLIGENT SYSTEM FOR TEXT AND IMAGE SPAM FILTERING Amol H. Malge 1, Dr. S. M. Chaware 2

More information

Proactive Drive Failure Prediction for Large Scale Storage Systems

Proactive Drive Failure Prediction for Large Scale Storage Systems Proactive Drive Failure Prediction for Large Scale Storage Systems Bingpeng Zhu, Gang Wang, Xiaoguang Liu 2, Dianming Hu 3, Sheng Lin, Jingwei Ma Nankai-Baidu Joint Lab, College of Information Technical

More information